Predicting Multi-Subsequent Events and Actors in Public Health Emergencies: An Event-Based Knowledge Graph Approach

Public health emergencies trigger a series of chain reactions with devastating impacts on society. In addition, subsequent events and participants in public health emergencies represent comprehensive scenarios of public health emergencies. Taking this information into account, predicting subsequent events and participants could motivate governments to take necessary countermeasures effectively. Therefore, we develop a prediction model of subsequent events and potential participants, i.e., subsequent multievent graph convolution network (SMEGCN), by utilizing events’ evolutionary information. Specifically, we take both relational information and semantic information into consideration to improve the prediction performance and simultaneously predict subsequent actors conveniently. Specifically, we collect data from five news datasets on the Sina microblog concerning the COVID-19 pandemic by employing a Python-based agent to test the performance of our model practically. The results show that embedding relational information, semantic information, and context inference into the prediction model could improve the model performance. Additionally, the comparative analysis indicates that the SMEGCN model is superior in predicting both subsequent events and participants. From the perspective of exemplifying analysis, social media, especially official media accounts, stimulates interactions between governments and the public and improves the management effectiveness of governments. The agile principle is applied to identify and handle a series of subsequent events and potential participants. Drawing conclusions from these discussions, the present study not only contributes to the literature on predicting events and participants theoretically and methodologically but also supplements practical suggestions for managing public health emergencies.